Google's self-driving vehicles are mastering complex situations on public roads, from cars going the wrong way to bicycles darting in front of traffic, as the technology company strives to win the high-profile race to achieve full vehicle automation, executives said on Wednesday.

The Alphabet Inc unit, which has been developing autonomous cars since 2009, said its self-driving vehicles had logged two million miles (3.2 million km) on public roads, and it continues to log about 25,000 miles of test drives per week.

Google has been a leader in testing autonomous vehicles on public roads, prompting traditional carmakers to step up their self-driving plans to avoid being eclipsed by the tech leader, which has yet to disclose a business strategy for its car project.

The Silicon Valley-based company is focused on making cars fully autonomous, with no need for a driver, which could make driving safer and more efficient and open up transportation to the disabled and aged. It said last year that such cars would be ready for production by 2020.

Its approach stands in sharp contrast to many carmakers, including Toyota Motor Corp and Tesla Motors Inc , which are moving towards autonomous driving in incremental steps, currently still requiring drivers.

Tesla has said it logged over 100 million miles since last October from drivers using its partially autonomous Autopilot system.

But miles driven on predictable highways are easier than navigating busy city streets, said Dmitri Dolgov, head of Google's self-driving technology effort.

Google's cars experience more complex situations than just following a car on the freeway, he said.

‘You have to have a deeper understanding of what's on the road and to the side,’ Dolgov told Reuters following a ride in a self-driving Lexus near Alphabet's Mountain View, California campus, during which the vehicle reacted to individuals in wheelchairs, pushing strollers and a car making a U-turn. ‘Every time you drive it's different.’

The Google car can detect whether a moving object is a child or a bicycle and anticipates that both can make fast, unpredictable movements. The car's driving system calculates the probability of such movements and uses the results to determining how the vehicle will react.

Those types of complicated social interactions are the last, most difficult element of autonomous vehicle technology, according to Dolgov.

‘You get to 90 percent autonomy in 10 percent of the time and then spend 90 percent of your time on the last 10 percent,’ he said.

The closest company to rival Google's self-driven miles in California is auto supplier Delphi Automotive Plc, with 16,662 autonomous miles, according to a January report filed with state regulators.